Benchmarking reports

from sklearn_benchmarks.report import Reporting
import pandas as pd
pd.set_option('display.max_colwidth', None)
pd.set_option('display.max_columns', None)
pd.set_option('display.max_rows', None)

reporting = Reporting(config_file_path="config.yml")
reporting.run()

Time report

	hour	min	sec
algo
KNeighborsClassifier	0.0	13.0	53.948621
daal4py_KNeighborsClassifier	0.0	3.0	26.714052
KNeighborsClassifier_kd_tree	0.0	6.0	56.286701
daal4py_KNeighborsClassifier_kd_tree	0.0	1.0	49.739476
KMeans_tall	0.0	1.0	42.166554
daal4py_KMeans_tall	0.0	1.0	18.936000
KMeans_short	0.0	0.0	19.896913
daal4py_KMeans_short	0.0	0.0	9.093467
LogisticRegression	0.0	1.0	3.360418
daal4py_LogisticRegression	0.0	0.0	55.243689
Ridge	0.0	0.0	48.074542
daal4py_Ridge	0.0	0.0	15.345816
total	0.0	32.0	38.884086

KNeighborsClassifier vs daal4py

Results

	estimator	function	mean_sklearn	stdev_sklearn	n_samples_train	n_samples	n_features	algorithm	n_jobs	n_neighbors	accuracy_score_sklearn	accuracy_score_daal4py	mean_daal4py	stdev_daal4py	speedup	stdev_speedup	profiling
0	KNeighborsClassifier	fit	0.149	0.002	1000000	1000000	100	brute	-1	1	NaN	NaN	0.498	0.001	0.299	0.003	See
1	KNeighborsClassifier	predict	0.167	0.014	1000000	1	100	brute	-1	1	1.0	0.0	0.085	0.001	1.954	0.162	See
2	KNeighborsClassifier	predict	25.681	1.056	1000000	1000	100	brute	-1	1	1.0	0.0	2.038	0.007	12.599	0.520	See
3	KNeighborsClassifier	fit	0.149	0.004	1000000	1000000	100	brute	-1	5	NaN	NaN	0.486	0.004	0.306	0.008	See
4	KNeighborsClassifier	predict	0.192	0.012	1000000	1	100	brute	-1	5	1.0	0.0	0.085	0.001	2.260	0.139	See
5	KNeighborsClassifier	predict	35.886	0.000	1000000	1000	100	brute	-1	5	1.0	0.0	2.028	0.008	17.696	0.068	See
6	KNeighborsClassifier	fit	0.134	0.000	1000000	1000000	100	brute	-1	100	NaN	NaN	0.484	0.001	0.277	0.001	See
7	KNeighborsClassifier	predict	0.187	0.016	1000000	1	100	brute	-1	100	1.0	0.0	0.085	0.000	2.187	0.183	See
8	KNeighborsClassifier	predict	35.764	0.000	1000000	1000	100	brute	-1	100	1.0	0.0	2.106	0.010	16.981	0.083	See
9	KNeighborsClassifier	fit	0.131	0.000	1000000	1000000	100	brute	1	1	NaN	NaN	0.486	0.003	0.270	0.002	See
10	KNeighborsClassifier	predict	0.188	0.001	1000000	1	100	brute	1	1	1.0	0.0	0.085	0.000	2.213	0.017	See
11	KNeighborsClassifier	predict	13.208	0.019	1000000	1000	100	brute	1	1	1.0	0.0	2.023	0.010	6.528	0.035	See
12	KNeighborsClassifier	fit	0.132	0.000	1000000	1000000	100	brute	1	5	NaN	NaN	0.484	0.001	0.272	0.001	See
13	KNeighborsClassifier	predict	0.203	0.002	1000000	1	100	brute	1	5	1.0	0.0	0.086	0.001	2.365	0.026	See
14	KNeighborsClassifier	predict	24.939	0.016	1000000	1000	100	brute	1	5	1.0	0.0	2.041	0.012	12.218	0.074	See
15	KNeighborsClassifier	fit	0.143	0.003	1000000	1000000	100	brute	1	100	NaN	NaN	0.485	0.004	0.295	0.006	See
16	KNeighborsClassifier	predict	0.207	0.003	1000000	1	100	brute	1	100	1.0	0.0	0.086	0.001	2.416	0.041	See
17	KNeighborsClassifier	predict	25.054	0.118	1000000	1000	100	brute	1	100	1.0	0.0	2.095	0.010	11.958	0.081	See
18	KNeighborsClassifier	fit	0.062	0.001	1000000	1000000	2	brute	-1	1	NaN	NaN	0.111	0.003	0.563	0.016	See
19	KNeighborsClassifier	predict	0.020	0.003	1000000	1	2	brute	-1	1	1.0	1.0	0.006	0.000	3.277	0.455	See
20	KNeighborsClassifier	predict	21.540	0.008	1000000	1000	2	brute	-1	1	1.0	1.0	0.317	0.002	67.891	0.443	See
21	KNeighborsClassifier	fit	0.061	0.000	1000000	1000000	2	brute	-1	5	NaN	NaN	0.110	0.003	0.556	0.016	See
22	KNeighborsClassifier	predict	0.030	0.001	1000000	1	2	brute	-1	5	1.0	1.0	0.007	0.001	4.494	0.621	See
23	KNeighborsClassifier	predict	31.879	0.000	1000000	1000	2	brute	-1	5	1.0	1.0	0.321	0.002	99.308	0.695	See
24	KNeighborsClassifier	fit	0.061	0.000	1000000	1000000	2	brute	-1	100	NaN	NaN	0.110	0.000	0.558	0.005	See
25	KNeighborsClassifier	predict	0.031	0.002	1000000	1	2	brute	-1	100	1.0	1.0	0.007	0.001	4.766	0.855	See
26	KNeighborsClassifier	predict	32.130	0.000	1000000	1000	2	brute	-1	100	1.0	1.0	0.379	0.004	84.700	0.843	See
27	KNeighborsClassifier	fit	0.063	0.000	1000000	1000000	2	brute	1	1	NaN	NaN	0.111	0.003	0.568	0.017	See
28	KNeighborsClassifier	predict	0.016	0.000	1000000	1	2	brute	1	1	1.0	1.0	0.006	0.001	2.448	0.217	See
29	KNeighborsClassifier	predict	10.563	0.009	1000000	1000	2	brute	1	1	1.0	1.0	0.323	0.008	32.682	0.837	See
30	KNeighborsClassifier	fit	0.062	0.000	1000000	1000000	2	brute	1	5	NaN	NaN	0.111	0.003	0.563	0.013	See
31	KNeighborsClassifier	predict	0.025	0.001	1000000	1	2	brute	1	5	1.0	1.0	0.006	0.000	4.073	0.266	See
32	KNeighborsClassifier	predict	21.199	0.008	1000000	1000	2	brute	1	5	1.0	1.0	0.319	0.004	66.366	0.731	See
33	KNeighborsClassifier	fit	0.062	0.000	1000000	1000000	2	brute	1	100	NaN	NaN	0.110	0.001	0.568	0.004	See
34	KNeighborsClassifier	predict	0.026	0.001	1000000	1	2	brute	1	100	1.0	1.0	0.006	0.000	4.054	0.340	See
35	KNeighborsClassifier	predict	21.448	0.017	1000000	1000	2	brute	1	100	1.0	1.0	0.382	0.003	56.175	0.438	See

Plots

Shared hyperparameters:

	value
algorithm	brute

KNeighborsClassifier_kd_tree vs daal4py

Results

	estimator	function	mean_sklearn	stdev_sklearn	n_samples_train	n_samples	n_features	algorithm	n_jobs	n_neighbors	accuracy_score_sklearn	accuracy_score_daal4py	mean_daal4py	stdev_daal4py	speedup	stdev_speedup	profiling
0	KNeighborsClassifier_kd_tree	fit	3.100	0.014	1000000	1000000	10	kd_tree	-1	1	NaN	NaN	0.752	0.013	4.124	0.076	See
1	KNeighborsClassifier_kd_tree	predict	0.004	0.001	1000000	1	10	kd_tree	-1	1	1.0	1.0	0.000	0.000	13.677	7.313	See
2	KNeighborsClassifier_kd_tree	predict	0.472	0.004	1000000	1000	10	kd_tree	-1	1	1.0	1.0	0.108	0.002	4.358	0.095	See
3	KNeighborsClassifier_kd_tree	fit	3.136	0.016	1000000	1000000	10	kd_tree	-1	5	NaN	NaN	0.771	0.005	4.067	0.033	See
4	KNeighborsClassifier_kd_tree	predict	0.004	0.001	1000000	1	10	kd_tree	-1	5	1.0	1.0	0.000	0.000	11.687	6.180	See
5	KNeighborsClassifier_kd_tree	predict	0.863	0.004	1000000	1000	10	kd_tree	-1	5	1.0	1.0	0.201	0.001	4.300	0.036	See
6	KNeighborsClassifier_kd_tree	fit	3.137	0.024	1000000	1000000	10	kd_tree	-1	100	NaN	NaN	0.787	0.023	3.986	0.120	See
7	KNeighborsClassifier_kd_tree	predict	0.011	0.001	1000000	1	10	kd_tree	-1	100	1.0	1.0	0.001	0.000	11.165	4.229	See
8	KNeighborsClassifier_kd_tree	predict	2.827	0.017	1000000	1000	10	kd_tree	-1	100	1.0	1.0	0.607	0.016	4.660	0.126	See
9	KNeighborsClassifier_kd_tree	fit	3.102	0.029	1000000	1000000	10	kd_tree	1	1	NaN	NaN	0.798	0.017	3.886	0.092	See
10	KNeighborsClassifier_kd_tree	predict	0.002	0.000	1000000	1	10	kd_tree	1	1	1.0	1.0	0.000	0.000	6.919	3.742	See
11	KNeighborsClassifier_kd_tree	predict	0.753	0.003	1000000	1000	10	kd_tree	1	1	1.0	1.0	0.115	0.001	6.555	0.086	See
12	KNeighborsClassifier_kd_tree	fit	3.133	0.033	1000000	1000000	10	kd_tree	1	5	NaN	NaN	0.752	0.017	4.167	0.106	See
13	KNeighborsClassifier_kd_tree	predict	0.003	0.001	1000000	1	10	kd_tree	1	5	1.0	1.0	0.000	0.000	7.907	3.650	See
14	KNeighborsClassifier_kd_tree	predict	1.452	0.010	1000000	1000	10	kd_tree	1	5	1.0	1.0	0.207	0.004	7.019	0.146	See
15	KNeighborsClassifier_kd_tree	fit	3.141	0.027	1000000	1000000	10	kd_tree	1	100	NaN	NaN	0.791	0.016	3.973	0.088	See
16	KNeighborsClassifier_kd_tree	predict	0.010	0.001	1000000	1	10	kd_tree	1	100	1.0	1.0	0.001	0.000	10.147	3.497	See
17	KNeighborsClassifier_kd_tree	predict	4.779	0.014	1000000	1000	10	kd_tree	1	100	1.0	1.0	0.633	0.004	7.546	0.057	See
18	KNeighborsClassifier_kd_tree	fit	1.419	0.006	1000000	1000000	2	kd_tree	-1	1	NaN	NaN	0.541	0.005	2.624	0.027	See
19	KNeighborsClassifier_kd_tree	predict	0.002	0.001	1000000	1	2	kd_tree	-1	1	1.0	0.0	0.000	0.000	18.977	14.735	See
20	KNeighborsClassifier_kd_tree	predict	0.030	0.001	1000000	1000	2	kd_tree	-1	1	1.0	0.0	0.001	0.000	34.822	14.526	See
21	KNeighborsClassifier_kd_tree	fit	1.433	0.011	1000000	1000000	2	kd_tree	-1	5	NaN	NaN	0.533	0.011	2.689	0.058	See
22	KNeighborsClassifier_kd_tree	predict	0.002	0.000	1000000	1	2	kd_tree	-1	5	1.0	0.0	0.000	0.000	17.744	12.446	See
23	KNeighborsClassifier_kd_tree	predict	0.030	0.000	1000000	1000	2	kd_tree	-1	5	1.0	0.0	0.001	0.001	21.488	8.282	See
24	KNeighborsClassifier_kd_tree	fit	1.351	0.006	1000000	1000000	2	kd_tree	-1	100	NaN	NaN	0.530	0.014	2.549	0.070	See
25	KNeighborsClassifier_kd_tree	predict	0.003	0.001	1000000	1	2	kd_tree	-1	100	1.0	0.0	0.000	0.000	20.580	14.655	See
26	KNeighborsClassifier_kd_tree	predict	0.053	0.004	1000000	1000	2	kd_tree	-1	100	1.0	0.0	0.008	0.001	6.782	1.113	See
27	KNeighborsClassifier_kd_tree	fit	1.370	0.016	1000000	1000000	2	kd_tree	1	1	NaN	NaN	0.536	0.013	2.559	0.070	See
28	KNeighborsClassifier_kd_tree	predict	0.001	0.000	1000000	1	2	kd_tree	1	1	1.0	0.0	0.000	0.000	5.927	4.724	See
29	KNeighborsClassifier_kd_tree	predict	0.027	0.000	1000000	1000	2	kd_tree	1	1	1.0	0.0	0.001	0.000	23.743	9.060	See
30	KNeighborsClassifier_kd_tree	fit	1.369	0.011	1000000	1000000	2	kd_tree	1	5	NaN	NaN	0.533	0.019	2.567	0.093	See
31	KNeighborsClassifier_kd_tree	predict	0.001	0.000	1000000	1	2	kd_tree	1	5	1.0	0.0	0.000	0.000	5.844	4.640	See
32	KNeighborsClassifier_kd_tree	predict	0.029	0.001	1000000	1000	2	kd_tree	1	5	1.0	0.0	0.001	0.000	24.543	6.325	See
33	KNeighborsClassifier_kd_tree	fit	1.360	0.008	1000000	1000000	2	kd_tree	1	100	NaN	NaN	0.540	0.007	2.517	0.037	See
34	KNeighborsClassifier_kd_tree	predict	0.001	0.000	1000000	1	2	kd_tree	1	100	1.0	0.0	0.000	0.000	5.784	4.346	See
35	KNeighborsClassifier_kd_tree	predict	0.058	0.001	1000000	1000	2	kd_tree	1	100	1.0	0.0	0.008	0.001	7.404	1.089	See

Plots

Shared hyperparameters:

	value
algorithm	kd_tree

KMeans_tall vs daal4py

Results

	estimator	function	mean_sklearn	stdev_sklearn	n_samples_train	n_samples	n_features	algorithm	init	max_iter	n_clusters	n_init	tol	n_iter_sklearn	adjusted_rand_score_sklearn	n_iter_daal4py	adjusted_rand_score_daal4py	mean_daal4py	stdev_daal4py	speedup	stdev_speedup	profiling
0	KMeans_tall	fit	0.649	0.011	1000000	1000000	2	full	k-means++	30	3	1	0.0	30.0	NaN	30.0	NaN	0.503	0.041	1.290	0.108	See
1	KMeans_tall	predict	0.000	0.000	1000000	1	2	full	k-means++	30	3	1	0.0	NaN	1.0	NaN	1.0	0.000	0.000	2.082	1.615	See
2	KMeans_tall	predict	0.000	0.000	1000000	1000	2	full	k-means++	30	3	1	0.0	NaN	1.0	NaN	1.0	0.000	0.000	1.681	1.038	See
3	KMeans_tall	fit	0.576	0.008	1000000	1000000	2	full	random	30	3	1	0.0	30.0	NaN	30.0	NaN	0.456	0.039	1.263	0.109	See
4	KMeans_tall	predict	0.000	0.000	1000000	1	2	full	random	30	3	1	0.0	NaN	1.0	NaN	1.0	0.000	0.000	2.087	1.521	See
5	KMeans_tall	predict	0.000	0.000	1000000	1000	2	full	random	30	3	1	0.0	NaN	1.0	NaN	1.0	0.000	0.000	1.817	1.248	See
6	KMeans_tall	fit	6.849	0.072	1000000	1000000	100	full	k-means++	30	3	1	0.0	30.0	NaN	30.0	NaN	3.136	0.028	2.184	0.030	See
7	KMeans_tall	predict	0.000	0.000	1000000	1	100	full	k-means++	30	3	1	0.0	NaN	1.0	NaN	1.0	0.000	0.000	1.707	1.381	See
8	KMeans_tall	predict	0.001	0.000	1000000	1000	100	full	k-means++	30	3	1	0.0	NaN	1.0	NaN	1.0	0.000	0.000	2.251	1.067	See
9	KMeans_tall	fit	6.232	0.061	1000000	1000000	100	full	random	30	3	1	0.0	30.0	NaN	30.0	NaN	2.937	0.018	2.121	0.024	See
10	KMeans_tall	predict	0.000	0.000	1000000	1	100	full	random	30	3	1	0.0	NaN	1.0	NaN	1.0	0.000	0.000	1.551	1.143	See
11	KMeans_tall	predict	0.001	0.000	1000000	1000	100	full	random	30	3	1	0.0	NaN	1.0	NaN	1.0	0.000	0.000	1.920	0.873	See

Plots

Shared hyperparameters:

	value
algorithm	full
n_clusters	3
max_iter	30
n_init	1
tol	0.0

KMeans_short vs daal4py

Results

	estimator	function	mean_sklearn	stdev_sklearn	n_samples_train	n_samples	n_features	algorithm	init	max_iter	n_clusters	n_init	tol	n_iter_sklearn	adjusted_rand_score_sklearn	n_iter_daal4py	adjusted_rand_score_daal4py	mean_daal4py	stdev_daal4py	speedup	stdev_speedup	profiling
0	KMeans_short	fit	0.349	0.030	10000	10000	2	full	k-means++	30	300	1	0.0	30.0	NaN	30.0	NaN	0.114	0.003	3.049	0.274	See
1	KMeans_short	predict	0.000	0.000	10000	1	2	full	k-means++	30	300	1	0.0	NaN	1.0	NaN	1.0	0.000	0.000	2.180	1.495	See
2	KMeans_short	predict	0.001	0.000	10000	1000	2	full	k-means++	30	300	1	0.0	NaN	1.0	NaN	1.0	0.001	0.000	1.282	0.343	See
3	KMeans_short	fit	0.130	0.001	10000	10000	2	full	random	30	300	1	0.0	30.0	NaN	30.0	NaN	0.050	0.001	2.606	0.061	See
4	KMeans_short	predict	0.000	0.000	10000	1	2	full	random	30	300	1	0.0	NaN	1.0	NaN	1.0	0.000	0.000	1.979	1.295	See
5	KMeans_short	predict	0.001	0.000	10000	1000	2	full	random	30	300	1	0.0	NaN	1.0	NaN	1.0	0.001	0.000	1.196	0.420	See
6	KMeans_short	fit	0.854	0.033	10000	10000	100	full	k-means++	30	300	1	0.0	22.0	NaN	22.0	NaN	0.401	0.015	2.131	0.114	See
7	KMeans_short	predict	0.000	0.000	10000	1	100	full	k-means++	30	300	1	0.0	NaN	1.0	NaN	1.0	0.000	0.000	1.634	0.997	See
8	KMeans_short	predict	0.007	0.002	10000	1000	100	full	k-means++	30	300	1	0.0	NaN	1.0	NaN	1.0	0.001	0.000	5.221	1.971	See
9	KMeans_short	fit	0.244	0.029	10000	10000	100	full	random	30	300	1	0.0	17.0	NaN	22.0	NaN	0.191	0.030	1.280	0.251	See
10	KMeans_short	predict	0.000	0.000	10000	1	100	full	random	30	300	1	0.0	NaN	1.0	NaN	1.0	0.000	0.000	1.743	0.971	See
11	KMeans_short	predict	0.007	0.003	10000	1000	100	full	random	30	300	1	0.0	NaN	1.0	NaN	1.0	0.001	0.000	5.417	2.055	See

Plots

Shared hyperparameters:

	value
algorithm	full
n_clusters	300
max_iter	30
n_init	1
tol	0.0

LogisticRegression vs daal4py

Results

	estimator	function	mean_sklearn	stdev_sklearn	n_samples_train	n_samples	n_features	C	class_weight	dual	fit_intercept	intercept_scaling	l1_ratio	max_iter	multi_class	n_jobs	penalty	random_state	solver	tol	verbose	warm_start	n_iter	mean_daal4py	stdev_daal4py	speedup	stdev_speedup	profiling
0	LogisticRegression	fit	11.631	0.039	1000000	1000000	100	1.0	NaN	False	True	1	NaN	100	auto	NaN	l2	NaN	lbfgs	0.0	0	False	[20]	11.322	0.023	1.027	0.004	See
1	LogisticRegression	predict	0.000	0.000	1000000	1	100	1.0	NaN	False	True	1	NaN	100	auto	NaN	l2	NaN	lbfgs	0.0	0	False	NaN	0.000	0.000	0.422	0.478	See
2	LogisticRegression	predict	0.000	0.000	1000000	1000	100	1.0	NaN	False	True	1	NaN	100	auto	NaN	l2	NaN	lbfgs	0.0	0	False	NaN	0.000	0.000	0.961	0.493	See
3	LogisticRegression	fit	0.790	0.014	1000	1000	10000	1.0	NaN	False	True	1	NaN	100	auto	NaN	l2	NaN	lbfgs	0.0	0	False	[27]	0.775	0.017	1.019	0.029	See
4	LogisticRegression	predict	0.000	0.000	1000	1	10000	1.0	NaN	False	True	1	NaN	100	auto	NaN	l2	NaN	lbfgs	0.0	0	False	NaN	0.001	0.000	0.132	0.086	See
5	LogisticRegression	predict	0.002	0.000	1000	100	10000	1.0	NaN	False	True	1	NaN	100	auto	NaN	l2	NaN	lbfgs	0.0	0	False	NaN	0.004	0.001	0.523	0.228	See

Plots

Shared hyperparameters:

	value
penalty	l2
dual	False
tol	0.0001
C	1.0
fit_intercept	True
intercept_scaling	1
class_weight	NaN
random_state	NaN
solver	lbfgs
max_iter	100
multi_class	auto
verbose	0
warm_start	False
n_jobs	NaN
l1_ratio	NaN

Ridge vs daal4py

Results

	estimator	function	mean_sklearn	stdev_sklearn	n_samples_train	n_samples	n_features	alpha	copy_X	fit_intercept	max_iter	normalize	random_state	solver	tol	n_iter	mean_daal4py	stdev_daal4py	speedup	stdev_speedup	profiling
0	Ridge	fit	1.775	0.046	100000	100000	1000	1.0	True	True	NaN	False	NaN	auto	0.001	NaN	0.959	0.005	1.851	0.049	See
1	Ridge	predict	0.000	0.000	100000	1	1000	1.0	True	True	NaN	False	NaN	auto	0.001	NaN	0.000	0.000	0.616	0.624	See
2	Ridge	predict	0.001	0.000	100000	1000	1000	1.0	True	True	NaN	False	NaN	auto	0.001	NaN	0.001	0.000	0.834	0.319	See
3	Ridge	fit	1.135	0.009	1000000	1000000	100	1.0	True	True	NaN	False	NaN	auto	0.001	NaN	0.245	0.002	4.636	0.052	See
4	Ridge	predict	0.000	0.000	1000000	1	100	1.0	True	True	NaN	False	NaN	auto	0.001	NaN	0.000	0.000	0.641	0.670	See
5	Ridge	predict	0.000	0.000	1000000	1000	100	1.0	True	True	NaN	False	NaN	auto	0.001	NaN	0.000	0.000	0.662	0.424	See

Plots

Shared hyperparameters:

	value
alpha	1.0
fit_intercept	True
normalize	False
copy_X	True
max_iter	NaN
tol	0.001
solver	auto
random_state	NaN

Environment information

{
  "system_info": {
    "python": "3.8.8 | packaged by conda-forge | (default, Feb 20 2021, 16:22:27)  [GCC 9.3.0]",
    "executable": "/usr/share/miniconda/envs/sklbench/bin/python",
    "machine": "Linux-5.4.0-1046-azure-x86_64-with-glibc2.10"
  },
  "dependencies_info": {
    "pip": "21.1",
    "setuptools": "49.6.0.post20210108",
    "sklearn": "0.24.1",
    "numpy": "1.20.2",
    "scipy": "1.6.2",
    "Cython": null,
    "pandas": "1.2.4",
    "matplotlib": null,
    "joblib": "1.0.1",
    "threadpoolctl": "2.1.0"
  },
  "threadpool_info": [
    {
      "filepath": "/usr/share/miniconda/envs/sklbench/lib/libopenblasp-r0.3.12.so",
      "prefix": "libopenblas",
      "user_api": "blas",
      "internal_api": "openblas",
      "version": "0.3.12",
      "num_threads": 2,
      "threading_layer": "pthreads"
    },
    {
      "filepath": "/usr/share/miniconda/envs/sklbench/lib/libgomp.so.1.0.0",
      "prefix": "libgomp",
      "user_api": "openmp",
      "internal_api": "openmp",
      "version": null,
      "num_threads": 2
    }
  ],
  "cpu_count": 2
}